11 Preflop GTO: Opening, 3-Bet/4-Bet & Defense Ranges
Preflop is the most-played street in poker. You make a preflop decision on every single hand you are dealt, and the range you enter the pot with constrains everything that happens afterward. Get your preflop ranges roughly right and your postflop game has a fighting chance; get them badly wrong and no amount of clever river play will rescue you. This chapter builds a game-theory-optimal (GTO) baseline for 100bb cash games (the equilibrium ranges that are hard to exploit) and explains the logic well enough that you can reconstruct a chart from principles when you have forgotten the exact cells.
By the end of this chapter you will be able to build and apply position-based preflop ranges, covering opening, 3-betting, 4-betting, flatting, and blind defense, for 100bb cash games, and reason from principles when memory fails.
A word on what “GTO preflop” means in practice. The ranges here approximate what modern preflop equilibrium solvers (MonkerSolver, GTO Wizard, Simple Preflop, and similar multiway tools) output for 100bb, standard-rake play without a HUD (heads-up display). Postflop solvers such as PioSOLVER are the right reference once you reach the flop, but they are not built to generate full multiway preflop equilibria, so use the preflop tools above for that. The ranges are not gospel: outputs shift with rake, antes, stack depth, and bet-sizing assumptions, and real opponents are rarely at equilibrium. Treat them as your default, the ranges you fall back to with no read and the reference point you deviate from once you have one.
11.1 Position Is the Master Variable
Every preflop range is a function of one dominant input: how many players act after you. The later your position, the more hands you can profitably play, because fewer opponents remain to wake up with a big hand and because you will more often have position postflop.
We use standard 6-max position names: UTG (under the gun, first to act), HJ (hijack), CO (cutoff), BTN (button), SB (small blind), BB (big blind). Full-ring (9-handed) inserts three extra early seats (UTG, UTG+1, UTG+2, the “early” block) plus a distinct lojack (LJ) between that block and the hijack, giving the full nine-seat order UTG, UTG+1, UTG+2, LJ, HJ, CO, BTN, SB, BB. The lojack is its own seat, one spot earlier than the hijack.
There is no single “good starting hand” list. A hand’s playability is entirely contextual. 7♠6♠ is a clear open on the BTN and a clear fold UTG at a full table. K♠J♠ is a solid but borderline UTG open (KJo only at low frequency) yet a marginal flat against a 3-bet. Stop memorizing hands and start memorizing positional ranges.
11.2 Raise-First-In (RFI) Ranges
When the pot is unopened and it is your turn, you either fold or raise. Limping (just calling the big blind) is essentially never part of an equilibrium RFI strategy, outside of specific SB constructions. Limping caps your range (leaves it without the strongest hands, which would have raised) and invites the BB to attack. Raise or fold.
Open sizing
- 6-max, 100bb online: Typical opens are 2.0–2.5bb from all positions, with many solver solutions favoring a flat 2.5bb or a position-dependent ramp (smaller late, larger early). A common simplification: open 2.5bb from every seat.
- Live cash: Because live tables call more and play more passively, 3bb to 4bb (or “pot”) opens are standard and correct: a larger bet charges the loose callers and builds a pot you have an equity edge in.
- From the SB: open larger, often 3bb, because you will be out of position the whole hand and want to discourage the BB’s wide, cheap defense.
The smaller the open, the wider you can profitably open (you risk less to win the blinds), but the more the blinds can defend. These two forces roughly cancel, which is why 2.0–2.5bb opens with appropriately wide ranges form a stable equilibrium online.
RFI by position (6-max, ~100bb)
The table below gives approximate opening frequencies and the character of each range. Frequencies are expressed as a percentage of all 1,326 starting combinations.
| Position | RFI % (approx.) | Character of the range |
|---|---|---|
| UTG | ~15–18% | Tight, value-dense: big pairs, big broadways, strong suited aces |
| HJ | ~19–22% | Adds more suited connectors and mid pairs |
| CO | ~26–30% | Wide; most suited hands, all pairs, many offsuit broadways |
| BTN | ~42–48% | Very wide; any pair, most suited, many offsuit gappers |
| SB | ~36–44% | Wide but raise-or-fold; a “complete/limp” range exists vs. some configs |
UTG (~15%) is the anchor everyone should know cold:
- All pairs 22+ (small pairs are a slight open, partly for set-mining, that is, calling cheaply to flop a set of three of a kind, and partly for balance)
- ATs+ (suited aces ATs and better), plus A5s–A2s as occasional balancing bluffs
- AJo+ (offsuit aces AJo and better)
- KTs+, QTs+, JTs, and the broadway offsuit hands KQo (sometimes KJo)
- Suited connectors roughly T9s–98s as the bottom of the suited bluffs
BTN (~45%) is the widest non-blind range. From the button you open essentially:
- Every pocket pair
- Every suited ace, every suited king down to about K5s, most suited queens and jacks
- All suited connectors and one-gappers down to about 54s
- A wide swath of offsuit broadways and offsuit aces (A2o+ in many solutions), K9o+, Q9o+, J9o, T9o
Opening too wide from early position is the single most common preflop leak in low-stakes games. A 25%+ UTG range looks aggressive and “modern,” but you will be out of position with too many dominated hands. When in doubt UTG, fold the offsuit junk and the weak suited gappers. Tight-and-aggressive from up front, loose-and-aggressive from the button.
Representative RFI grids
Prose hand-lists are fine for intuition, but to actually drill a range you need the grid. Read every grid in this chapter as the standard 13x13 hand matrix: rows and columns run A, K, Q, J, T, 9, 8, 7, 6, 5, 4, 3, 2; the diagonal is the pocket pairs, cells above and right of it are suited, and cells below and left are offsuit. For the RFI grids: R = open (raise-first-in), r = mixed / low-frequency open, · = fold.
The 13x13 grids in this chapter are constructed reference ranges, hand-built to sit very close to modern preflop-solver output but deliberately rounded into clean, memorizable cells rather than transcribed literally from any one solve. A true equilibrium is messier than a single letter per cell can show. Many of the boundary cells are rake- and open-size-dependent, and a real solution plays them as mixes, opening (or 3-betting) some fraction of the time and folding the rest, not the pure action the letter implies. So read the edges as “roughly a coin-flip, lean this way,” not as law. For instance, J8s and T8s as CO opens, and 22 as a UTG open at 100bb, are close, frequency-dependent calls that tip toward folding under heavier rake or a larger open size; they are charted here as opens because that is the cleaner default to memorize, not because they are hard rules. Treat the dense, obviously-strong cells (the pairs, the broadways, the strong suited aces) as fixed, and the marginal cells at the perimeter as negotiable.
UTG open (6-max, ~15%). The value-dense anchor: big pairs and broadways, strong suited aces, the wheel suited aces as balancing bluffs, and only the very top suited connectors.
| A | K | Q | J | T | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | R | R | R | R | R | · | · | · | · | r | r | r | r |
| K | R | R | R | R | R | · | · | · | · | · | · | · | · |
| Q | R | R | R | R | R | · | · | · | · | · | · | · | · |
| J | R | r | · | R | R | · | · | · | · | · | · | · | · |
| T | · | · | · | · | R | R | · | · | · | · | · | · | · |
| 9 | · | · | · | · | · | R | R | · | · | · | · | · | · |
| 8 | · | · | · | · | · | · | R | · | · | · | · | · | · |
| 7 | · | · | · | · | · | · | · | R | · | · | · | · | · |
| 6 | · | · | · | · | · | · | · | · | R | · | · | · | · |
| 5 | · | · | · | · | · | · | · | · | · | R | · | · | · |
| 4 | · | · | · | · | · | · | · | · | · | · | R | · | · |
| 3 | · | · | · | · | · | · | · | · | · | · | · | R | · |
| 2 | · | · | · | · | · | · | · | · | · | · | · | · | R |
BTN open (6-max, ~45%). The widest non-blind range: every pair, every suited ace, most suited kings/queens/jacks, the full ladder of suited connectors and one-gappers down to the 50s, and a deep block of offsuit aces and broadways (the offsuit wheel aces and weakest suited Kx/Qx as mixes).
| A | K | Q | J | T | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | R | R | R | R | R | R | R | R | R | R | R | R | R |
| K | R | R | R | R | R | R | R | R | R | R | r | r | r |
| Q | R | R | R | R | R | R | R | R | R | R | r | r | · |
| J | R | R | R | R | R | R | R | R | r | r | · | · | · |
| T | R | R | R | R | R | R | R | R | r | · | · | · | · |
| 9 | R | R | R | R | R | R | R | R | r | · | · | · | · |
| 8 | R | r | · | · | · | r | R | R | R | r | · | · | · |
| 7 | r | · | · | · | · | · | · | R | R | R | · | · | · |
| 6 | r | · | · | · | · | · | · | · | R | R | r | · | · |
| 5 | r | · | · | · | · | · | · | · | · | R | R | r | · |
| 4 | r | · | · | · | · | · | · | · | · | · | R | r | · |
| 3 | r | · | · | · | · | · | · | · | · | · | · | R | · |
| 2 | r | · | · | · | · | · | · | · | · | · | · | · | R |
CO open (6-max, ~27%). The middle of the RFI ladder and the range used in this chapter’s worked example. Wider than UTG but not yet button-wide: every pair, every suited ace, suited kings down to the mid-cards, the broadway-to-mid suited queens/jacks/tens, the full block of suited connectors and the strongest one-gappers, plus a healthy slab of offsuit broadways and offsuit aces. The marginal Kx/Qx suited, the weak offsuit broadways, and the lower one-gappers come in as mixes.
| A | K | Q | J | T | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | R | R | R | R | R | R | R | R | R | R | R | R | R |
| K | R | R | R | R | R | R | R | R | r | r | · | · | · |
| Q | R | R | R | R | R | R | R | r | · | · | · | · | · |
| J | R | R | r | R | R | R | R | r | · | · | · | · | · |
| T | R | R | r | r | R | R | R | r | · | · | · | · | · |
| 9 | R | r | · | · | · | R | R | R | r | · | · | · | · |
| 8 | r | · | · | · | · | · | R | R | R | · | · | · | · |
| 7 | · | · | · | · | · | · | · | R | R | r | · | · | · |
| 6 | · | · | · | · | · | · | · | · | R | R | r | · | · |
| 5 | · | · | · | · | · | · | · | · | · | R | R | · | · |
| 4 | · | · | · | · | · | · | · | · | · | · | R | r | · |
| 3 | · | · | · | · | · | · | · | · | · | · | · | R | · |
| 2 | · | · | · | · | · | · | · | · | · | · | · | · | R |
That CO grid works out to roughly 27% of all combinations once the r mixes are weighted, squarely in the ~26–30% band above. The HJ (~19–22%) and SB ranges are charted only in prose by design. The HJ sits between the UTG and CO grids (start from CO and shave the weakest suited gappers, the loosest offsuit broadways, and a couple of the marginal suited Kx/Qx mixes), while the SB is a special raise-or-fold case that also folds in a small completing range against some configurations rather than running a clean RFI. When you need an exact HJ or SB chart, interpolate from the two anchors here (UTG and BTN) and the CO grid between them.
Full-ring differences
Full-ring (9-max) is tighter in every early seat because there are simply more players left to act. The four seats ahead of the hijack (UTG, UTG+1, UTG+2, and the lojack) each have one fewer opponent behind them than the seat before, so each opens a notch wider than the last. Rough RFI frequencies:
- UTG (~10–13%): eight players behind, the tightest seat at the table; basically the 6-max UTG range minus the speculative suited connectors and weakest suited aces.
- UTG+1 (~12–14%): seven behind, a touch wider now that one early seat sits behind you instead of in front.
- UTG+2 (~13–16%): six behind, wider still but a hair tighter than a 6-max UTG.
- LJ (~15–18%): five players behind, the same count as a 6-max UTG, so the lojack plays like the 6-max UTG anchor.
From the hijack onward the number of players left to act is identical in both formats (HJ has four behind, CO three, BTN two, in 6-max and 9-max alike), so those seats converge with their 6-max counterparts. The practical rule: in full ring, the three early seats each play one fold-tier tighter than a 6-max UTG, the lojack plays the 6-max UTG range itself, then you run ordinary 6-max ranges from the hijack down. The trap in the old “just play 6-max from the hijack on” shorthand is that it tempts you to open the lojack like a hijack, far too wide for a seat that still has five players left to act.
Opening ranges are not the only thing that tightens. With more live ranges yet to act, your 3-bet, squeeze, and cold-call frequencies all shrink too: more hands behind can wake up with a premium, so light 3-bets get punished more often, squeezes run into more legitimate continues, and cold-calls more frequently leave you sandwiched. Treat full-ring as 6-max with the aggression dialed back a notch wherever there are extra players left to act.
11.3 3-Betting: Linear vs. Polarized
When someone opens and you want to re-raise, your 3-bet range takes one of two shapes:
- Linear (merged): all of your strongest hands. You 3-bet the top of your range and flat or fold the rest. This shape fits when the caller behind you is weak or passive, when you are up against a very wide opener you dominate, and specifically from the small blind, where flatting is bad because the big blind acts behind you and can squeeze or overcall, so you cannot profitably keep a flatting range.
- Polarized: your strongest value hands plus a selection of weak-but-playable bluffs (often suited wheel aces and suited connectors or gappers), while hands of middling strength are flatted instead. This fits in position against a reasonable opener, where flatting your medium hands realizes equity better than bloating the pot with them. The big blind also 3-bets polarized: even though it is out of position, it closes the action and gets a price to flat its medium hands, so it splits into a polar 3-bet (value plus bluffs) on top of a very wide flat. The contrast with the SB is the key one: the SB linearizes because it cannot flat well, while the BB polarizes because it can.
Why polarize in position?
Imagine the CO opens and you are on the BTN with K♥Q♠. The hand is too strong to fold but plays beautifully as a flat: you keep dominated hands like KJo and QJo in the opener’s continuing range and you realize equity with position. If instead you held A♠5♠ (a nut blocker, decent equity when called, but poor flatting value), it makes a far better 3-bet bluff. So your BTN 3-bet range becomes polarized: AA–QQ, AK, AQs for value at the top; A5s–A2s and K5s as blocker bluffs at the bottom; and the broadways, connectors, and medium pairs in the middle go into the flatting range.
Here is that BTN polar 3-bet against a CO open as a grid. Codes: V = 3-bet for value, B = 3-bet bluff, C = flat/call, · = fold. Notice the shape: the two ends (V at the top, B at the bottom) re-raise while the entire connected, broadway-heavy middle flats (C). That gap is what “polarized” means.
| A | K | Q | J | T | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | V | V | V | C | C | C | C | C | C | B | B | B | B |
| K | V | V | C | C | C | C | C | C | C | B | · | · | · |
| Q | C | C | V | C | C | C | C | · | · | · | · | · | · |
| J | C | C | · | V | C | C | C | · | · | · | · | · | · |
| T | C | · | · | · | C | C | C | · | · | · | · | · | · |
| 9 | · | · | · | · | · | C | C | C | · | · | · | · | · |
| 8 | · | · | · | · | · | · | C | C | C | · | · | · | · |
| 7 | · | · | · | · | · | · | · | C | C | C | · | · | · |
| 6 | · | · | · | · | · | · | · | · | C | C | C | · | · |
| 5 | · | · | · | · | · | · | · | · | · | C | C | · | · |
| 4 | · | · | · | · | · | · | · | · | · | · | C | · | · |
| 3 | · | · | · | · | · | · | · | · | · | · | · | C | · |
| 2 | · | · | · | · | · | · | · | · | · | · | · | · | C |
(JJ and AQo are shown at their dominant action but are genuine 3-bet/flat mixes: JJ leans value 3-bet, AQo leans flat, and both legitimately do some of each. QQ never flats here; it is a pure value 3-bet, as discussed under flatting below.)
Why linearize from the small blind?
From the SB facing a BTN open, you have no position and flatting is dangerous (out of position with a capped range, and the BB can squeeze). So the SB’s main weapon is a linear/merged 3-bet: 3-bet a dense block of your best hands (pairs 55+, ATs+, KQs, AJo+, KQo type hands) and fold most of the rest, mixing in some suited bluffs for balance. Solvers do flat some hands from the SB, but the strategy leans far more toward 3-bet-or-fold than the in-position equivalent. The big blind is the opposite case, covered in “Big-Blind Defense Is Wide” below.
3-bet sizing
- In position (e.g., BTN vs CO): smaller, around 3x the open (open 2.5bb → 3-bet ~7.5–8bb). Position lets you use a smaller raise.
- Out of position (e.g., BB vs BTN, SB vs CO): larger, around 4x the open (open 2.5bb → 3-bet ~10–11bb), because you need to deny the opener cheap, in-position equity realization.
- Live: scale up for larger opens and stickier opponents; against a 4bb open, a 3-bet to ~12bb in position or ~14bb out of position is reasonable.
How often must a 3-bet bluff work? Run the same break-even arithmetic you use for steals and BB defense. A BTN 3-bet to 8bb over a 2.5bb CO open risks 8bb to win the ~4bb already in the middle (2.5 open + 1.0 BB + 0.5 SB), so as a pure bluff (treated as having zero equity when called) it needs the opener to fold 8 / (8 + 4) = 67% of the time to show an immediate profit. No opener folds two-thirds of a legitimate opening range to a single 3-bet, which is exactly why the bluff cannot rely on fold equity (the profit from forcing folds) alone. The nut blockers and the postflop playability of a hand like A5s make up the rest, turning a play that loses on fold equity in isolation into a profitable one overall.
3-bet bluffs are not random trash. They are chosen for two properties: (1) card removal (blockers), where A♠5♠ blocks AA, AK, and AQ and reduces the chance the opener holds a 4-bettable hand, and (2) playability when called, since suited wheel aces and suited connectors flop straights, flushes, and pairs with draws. A♣7♥2♦ has neither property, so it is never a 3-bet bluff; A♠5♠ has both.
11.4 Facing a 3-bet (as the Opener)
You opened, someone 3-bet, and now you have three choices: fold, flat (call the 3-bet), or 4-bet. This is a high-frequency core decision, since every wide RFI guarantees you will often be the one getting re-raised, and whether you keep a flatting range at all depends almost entirely on position.
In position (you opened and a later seat 3-bet you, e.g., CO opens, BTN 3-bets and the CO continues; or BTN opens, a blind 3-bets and the BTN acts last postflop): you keep a genuine flatting range. Because you will have position on every street and are getting a reasonable price, you call the hands that flop well and would hate to be 4-bet off their equity, rather than collapsing into pure 4-bet-or-fold:
- Pocket pairs that flop sets: roughly 22–JJ, calling cheaply to set-mine against a 3-bet range that has already capped itself (the biggest pairs mix in some 4-bets instead).
- Suited broadways that dominate and are dominated: AQs, AJs, KQs, KJs, QJs, JTs, which flop pairs, draws, and second-best hands you can stack the opener’s bluffs against.
- Suited connectors and gappers with playability: T9s–65s and a few suited one-gappers, which realize equity well in position.
On top of that flat you 4-bet the top of your range for value (QQ+, AK, with AQs and JJ going either way by configuration) plus a few blocker bluffs (A5s, A4s type), and you fold the offsuit junk and the dominated weak aces and broadways that neither flat nor 4-bet well. Note the contrast with KJ from the opening key idea: a fine open, yet facing a 3-bet only a marginal flat, the kind of dominated broadway you fold more often than you continue.
Out of position (you opened from an early or middle seat and a later seat 3-bet you): flatting is far more dangerous, because you will act first on every street with a capped range while the 3-bettor has position and initiative. So you skew hard toward 4-bet-or-fold, keeping only a thin, mostly suited flatting sliver (if any) and pushing most of your continues into a tight value-plus-blocker-bluff 4-bet. The deeper the stacks and the more clearly the 3-bettor is in position, the more of that marginal flatting region collapses into folds and 4-bets.
The 4-bet structure itself (the value blocks, how bluffs are chosen, and sizing) is detailed in the next section.
11.5 4-Betting, 5-Betting & Jamming
When your 3-bet gets 4-bet, or you face a 4-bet after opening, stacks start to commit and ranges get very narrow and value-heavy.
4-bet structure
At 100bb, a 4-bet is itself polarized: a block of value hands that want the money in (roughly QQ+, AK, sometimes JJ or AQs depending on positions) plus a few bluff 4-bets chosen, again, for blockers. The canonical 4-bet bluff is A5s (or A4s, A5o), because the ace blocks AA and AK, the two hands most likely to continue against your 4-bet. Sizing at 100bb is typically about 2.2–2.5x the 3-bet out of position and a bit smaller in position, large enough to set up a stack-off but not so large that you forfeit fold equity.
The commitment math
Once a 4-bet goes in, the pot is large relative to the remaining stacks. A hand like AK or KK facing a 4-bet usually wants to either 5-bet jam or call committed. AA and KK are happy to get 100bb in preflop, while QQ is closer to indifferent and only does so at low frequency (against equilibrium 4-bet/5-bet ranges it is dominated by KK and AA and merely flips with AK). The 5-bet at 100bb is almost always all-in (a jam), with rarely room for a non-all-in 5-bet, and the jamming range is extremely tight: predominantly AA, KK, AK, with QQ and the A5s-type bluffs mixed in at low frequency for balance.
Stacking off light against a 4-bet/5-bet war. At equilibrium, the player who 4-bets, and especially the one who 5-bet jams, is showing a range so strong that calling off with JJ, AQ, or even TT is often a losing play against tight opponents. Against unknown but solid regulars, treat a 4-bet as roughly “QQ+/AK with a few bluffs” and a 5-bet jam as “KK+/AK”, close to AA-only against the very tightest opponents. Fold your bluff-catchers unless you have a specific read that they are over-bluffing.
11.6 Flatting (Cold-Calling) Ranges
Flatting an open (calling rather than 3-betting) is mostly an in-position tool. The strongest flatting spot in poker is the BTN versus a CO or HJ open: you have position, the pot stays small, and you keep the opener’s dominated hands in. A representative BTN flatting range versus a CO open includes:
- Mid pairs that don’t want to 3-bet for value or get 4-bet off their equity: roughly 22–TT (with JJ and AQo sometimes flatting, sometimes 3-betting as a mix). QQ is a pure value 3-bet here. Flatting a premium that wants to 3-bet for value and deny equity, only to play a bloated pot in the dark, is a clear leak.
- Suited broadways that dominate and are dominated: AJs, ATs, KJs, KTs, QJs, JTs
- KQo, AQo, AJo type hands that prefer to keep the pot small
- Suited connectors T9s–65s for their postflop playability
Cold-calling out of position (in the blinds or sandwiched between players) is dangerous and largely avoided in solver play, with one big exception covered next: BB defense. When you do flat out of position, your range gets run over, because you are capped (you would have 3-bet your best hands) and have to act first on every street.
“Cold-calling in the small blind.” Flatting an open from the SB invites the BB to come along for a cheap price (or squeeze you), and leaves you out of position against two opponents with a capped range. Against most opens, the SB should be 3-bet-or-fold. Only construct a SB flatting range against very wide late-position opens, and even then keep it small and connected.
11.7 Big-Blind Defense Is Wide
The big blind is the most-defended seat in poker for one structural reason: you have already invested one big blind and you are getting a discount to continue. Facing a 2.5bb open, you only need to add 1.5bb to contest a pot that already contains 4.0bb (2.5 open + 1.0 BB + 0.5 SB), so you are getting about 2.67-to-1 immediate odds and need roughly 27% raw equity to call (1.5 / (4.0 + 1.5) = 1.5/5.5 ≈ 27.3%). Closing the action means no one can raise behind you, which loosens the requirement further. (These no-ante numbers assume the 100bb cash structure used throughout this chapter; a big-blind ante would lower the threshold further, but there is none here.)
Consequently, BB calling ranges are enormous: against a small BTN open the BB can correctly defend (call or 3-bet) a bit under half of all hands, roughly ~44–48% by combination frequency. (The “50%+” figure you often hear is real, but it belongs to a 2bb min-open; a 2.5bb open like the one here prices the BB out of the worst offsuit junk and pulls the defend toward ~44%.) Tie the boundary to the open size. Against the 2.5bb open used throughout this chapter, the defending region reaches down to hands like K8o, Q9o, J8o, and 84s, the offsuit and weak-suited hands that still flop enough equity to justify the price. The truly trashy stuff (K4o, Q6o, J7o, 83s) defends only against a 2bb min-open; against the 2.5bb open those become folds or tiny mixes, because you are paying more for the same hand. The wider and smaller the open, the wider the BB defends; raise the open size and the boundary marches back up.
There is a standard way to anchor “how much of my range must continue” to a number: minimum defense frequency (MDF) and its complement alpha. MDF = pot / (pot + bet) is the fraction the field must defend so that a zero-equity bluff cannot auto-profit; alpha = bet / (pot + bet) is the maximum fraction it may fold. Treat the open as the opener risking 2.5bb to win the 1.5bb of blinds already out there. The steal breaks even as a pure bluff when it succeeds 2.5 / (2.5 + 1.5) = 62.5% of the time, so the blinds must collectively defend at least 1.5 / (1.5 + 2.5) = 37.5% (alpha ≈ 62.5%). Because the SB usually folds, the BB carries almost all of that load alone. In practice the BB defends more than this 37.5% floor (the ~44–48% noted above), because the direct price demands only ~27% equity and most hands clear that bar easily. It does not defend everything because out of position you realize less of your raw equity than the pot odds imply (the equity-realization penalty in the Key idea below), so the practical continue threshold sits above the naive MDF floor.
Two crucial caveats:
- Wide does not mean fit-or-fold. Defending K4o profitably requires you to realize equity postflop: float flops, bluff-catch correctly, and not fold every time you whiff. If you defend wide and then surrender every pot you do not smash, you simply hand the money back.
- Defense width collapses as the open grows larger and as positions get earlier. Versus a 4bb live UTG open, the BB’s correct defending range is far tighter: you are getting a worse price against a stronger range and you will be out of position. Do not defend 83s against a UTG raise just because “BB defense is wide.”
Your equity requirement to call in the BB falls out of the pot odds, but your realization of that equity is discounted by being out of position. Solvers bake this “equity realization” penalty into the ranges, which is why the BB still folds a meaningful chunk of hands despite the great price. The price tells you the floor; position pulls the practical threshold back up.
Here is a representative BB defense grid against a 2.5bb BTN open (SB folded, so you close the action). Codes: 3 = 3-bet (the polar value and bluff cells), C = call/flat, · = fold. Several C cells also mix in a low-frequency 3-bet for balance (e.g., suited Kx, suited connectors); the grid shows their dominant action. Notice how the defending region (everything not ·) covers well over half the cells of the matrix, about 57% (96 of 169 cells), yet by combination frequency the BB defends only about 44% of hands (586 of 1,326 combos). The two figures diverge because the fold region is concentrated in the high-combo offsuit cells (12 combos each), while the defends cluster in the low-combo suited cells (4 combos) and pocket pairs (6 combos), so counting cells overstates how often you actually continue. Defend by the grid rather than by “half the chart”: literally calling 50%+ of combos here would over-defend the worst offsuit junk by several points of range and bleed EV. The 3-bets sit at the two ends (premiums plus the wheel suited aces), never in the flatted middle.
| A | K | Q | J | T | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | 3 | 3 | 3 | C | C | C | C | C | C | 3 | 3 | 3 | C |
| K | 3 | 3 | C | C | C | C | C | C | C | C | C | C | C |
| Q | C | C | 3 | C | C | C | C | C | C | C | C | C | C |
| J | C | C | C | 3 | C | C | C | C | C | C | C | · | · |
| T | C | C | C | C | C | C | C | C | C | C | · | · | · |
| 9 | C | C | C | C | C | C | C | C | C | C | · | · | · |
| 8 | C | C | · | C | C | C | C | C | C | C | C | · | · |
| 7 | · | · | · | · | · | · | C | C | C | C | C | · | · |
| 6 | · | · | · | · | · | · | · | · | C | C | C | · | · |
| 5 | C | · | · | · | · | · | · | · | · | C | C | C | · |
| 4 | · | · | · | · | · | · | · | · | · | · | C | C | · |
| 3 | · | · | · | · | · | · | · | · | · | · | · | C | · |
| 2 | · | · | · | · | · | · | · | · | · | · | · | · | C |
Read the offsuit boundary off the lower-left triangle: K8o, Q9o, J8o defend while K4o, Q6o, J7o fold, matching the open-size boundary above.
The BB also has a robust 3-bet range, but unlike the SB it is polarized. Because the BB closes the action and gets such a good price, it flats its medium hands rather than 3-betting them, so its 3-bet splits into two ends: premium value (roughly QQ+, AK, AQs, and the strongest pairs) at the top, plus a chosen set of bluffs (A3s–A5s, the visible wheel-ace 3-bets in the grid, with A2s mixing mostly call, some Kxs and suited connectors) at the bottom. The middle (broadways, suited gappers, small and medium pairs) goes into the wide flat. The 3-bet must contain bluffs for balance: a value-only BB 3-bet is trivially exploitable, since opponents fold everything but the nuts to it. Size it large, around ~10–11bb vs a 2.5bb open (≈4–4.5x), to deny the opener cheap, in-position equity realization.
11.8 The Squeeze
A squeeze is a 3-bet made after an open and one or more callers. It is one of the most profitable and under-used preflop plays, and it has special mechanics:
- You get extra dead money. The caller(s) have put chips in without the range strength to continue against a big raise. That dead money improves your immediate price on a bluff.
- You pressure the “sandwiched” caller, who now faces a 3-bet with a capped flatting range (they didn’t 3-bet, so they don’t have the nuts) and players still to act behind.
- Size up. With more players and more dead money, squeeze larger than a heads-up 3-bet. A common rule is the base 3-bet size plus about one extra opener-raise per caller. Starting from the out-of-position base of ~10–11bb (used everywhere in this chapter), open 2.5bb plus one caller gives a squeeze to roughly 13bb (the 10–11bb base plus one extra 2.5bb open).
Squeeze ranges are polarized: strong value (QQ+, AK, AQs) that wants to play a big pot against the capped callers, plus blocker-heavy bluffs (A5s, KQs, suited wheel aces). Avoid squeezing the medium hands, which would rather overcall and see a flop multiway.
Deal yourself 20 hands and, before looking at any chart, classify each one in a fixed spot, say “it folds to me on the BTN.” For each hand say out loud: open or fold? Then check against the BTN RFI range above. Repeat for “UTG to act” and “BB facing a 2.5bb BTN open.” Do this until your snap judgment matches the chart 18 out of 20 times. Speed and accuracy on these defaults free your mind for the reads that actually win money.
11.9 A Fully Worked Example
Setup: 6-max, 100bb effective, online, standard rake. Folds to the CO, who opens to 2.5bb. The BTN (a solid, aggressive regular) calls. You are in the BB with A♠5♠. The SB folds. Action is on you.
Step 1. Identify the spot. This is a squeeze spot: an open plus one cold-caller, and you are in the BB closing the action. There is dead money in the SB’s fold and in the BTN’s flat.
Step 2. Classify the hand. A♠5♠ is a premier squeeze bluff. The ace blocks AA and AK (the hands most likely to 4-bet or call strongly), the 5 gives wheel and straight potential, and the suit gives flush potential. It is a poor pure flat (dominated-ace problems, out of position, multiway) but an excellent polarized bluff.
Step 3. Choose the action and size. Squeeze. With three other live ranges and dead money, size up: the base out-of-position 3-bet of ~10–11bb plus roughly one opener-size (2.5bb) for the caller, a squeeze to about 13bb.
Step 4. Reason through the responses.
- The CO opened a ~27% range from the cutoff. Against a large BB squeeze, most of that range (suited gappers, offsuit broadways, small pairs) has to fold. The CO continues with roughly QQ+, AK, AQs, and a few bluff 4-bets or strong flats. You hold a card-removal advantage: your ace makes their AA and AK less likely.
- The BTN flatted, which caps their range: they almost certainly do not have AA or KK (those would usually 3-bet themselves). Facing your squeeze plus a still-live opener, the BTN’s flat-behind range is fragile and folds often.
- When everyone folds, you scoop the dead money immediately, the primary way this hand profits.
- When you do get called or 4-bet, A♠5♠ still flops reasonably, and your ace-blocker means the 4-bets you face are weighted toward genuine nut hands (fold) rather than bluffs.
Step 5. The takeaway. You squeeze A♠5♠ not for its strength but because the structure (dead money, a capped caller, your blockers) makes a polarized re-raise more profitable than folding or flatting. Swap your cards for K♥Q♠ and the picture changes: in a different, in-position spot you might flat K♥Q♠, but here you are out of position closing the action, so folding is fine, with an occasional overcall defensible given the price and the multiway pot you close. What you should not do with K♥Q♠ here is squeeze it as a bluff, since it has no nut blocker and is the kind of medium hand that prefers to stay out of a re-raised pot. Swap to Q♦Q♣ and you squeeze for value at the same size. Same spot, three hand classes, three reasons.
11.10 Putting It Together: A Reference Hierarchy
When you sit down at 100bb, carry this mental scaffolding:
- Unopened pot: raise or fold per your positional RFI. Tight UTG, very wide by the BTN. Never limp.
- One raise in front, you’re in position: polarized 3-bet (value plus blocker bluffs), or flat your medium hands, or fold the rest.
- One raise in front, you’re out of position (blinds): from the SB, lean toward a linear/merged 3-bet-or-fold (you can’t flat well). From the BB, defend very wide by calling because of the price, and 3-bet a polarized range (premium value plus blocker bluffs) on top of that flat, never a value-only 3-bet.
- Raise plus caller(s): consider the polarized squeeze, sized up for the dead money.
- You opened and got 3-bet: in position, keep a real flatting range (pairs that flop sets, suited broadways like AQs, KQs, AJs, suited connectors) and 4-bet QQ+/AK plus A5s-type blocker bluffs, folding the rest. Out of position, skew to 4-bet-or-fold with little or no flatting.
- Facing a 4-bet: narrow to value (QQ+, AK) plus a sprinkle of A5s-type bluffs; everything in between folds.
- Facing a 5-bet jam at 100bb: continue only with the very top, roughly KK+/AK, unless you have a concrete read that they are jamming light.
These are your defaults, the equilibrium you return to when the table gives you nothing to exploit. The next chapters take these baselines and bend them: when a player opens too wide, you 3-bet wider and call down lighter; when a player only 4-bets the nuts, you fold every bluff-catcher and re-bluff relentlessly. You cannot deviate intelligently from a baseline you do not own. Drill these ranges until they are automatic, and the thinking part of your brain is freed for the part of poker that actually beats other humans: reading them.